Novelty Test for Sub-Ranges

D. Visser (NL)

1 Introduction

The EPO uses a special novelty test to determine whether or not a claimed sub-range of a broader, known range is novel over that known range. According to this test, a sub-range must comply with each of the three criteria of the test to be novel.

Any novelty test must be based on the novelty requirement of Art. 54(1) EPC:

“An invention shall be considered to be new if it does not form part of the state of the art.”

The state of the art is defined in Art. 54(2):

“The state of the art shall be held to comprise everything made available to the public by means of a written or oral description, by use, or in any other way, before the date of filing of the European patent application.”

The word ‘available’ emphasises the possibility that the public can take note of the invention, according to the Travaux Préparatoires EPC 1973[1]. The extent of the disclosure of the prior art is determined by the disclosure test: the subject-matter which the person skilled in the art can derive directly and unambiguously, using common general knowledge, from the prior art[2].
Recent developments in case law have given rise to uncertainty in the interpretation of the novelty test for sub-ranges, in particular about what information the prior art makes available. Moreover, differences in interpretation between departments of first and second instance have emerged, recently resulting in the Guidelines dropping one of the criteria while the book Case Law of the Boards of Appeal maintains the three criteria. The differences cause legal uncertainty for users of the EPC.

This article will discuss the developments and differences in interpretation for each of the three criteria, with an emphasis on what subject-matter a broad parameter range discloses and the effect of this disclosure on the novelty assessment of a claimed sub-range of the known broad range. It concludes with a plea to refer the novelty test for sub-ranges to the Enlarged Board of Appeal to bring to an end years of legal uncertainty.[3]

The novelty tests for overlapping ranges and for purity ranges of a compound[4] use similar concepts as the novelty test for sub-ranges and also suffer from diverging case law. Several arguments developed in this article apply to these novelty tests as well. A referral should therefore consider also these novelty tests.

2 Origin of the novelty test for sub-ranges

2.1 First decision on novelty of sub-ranges

T198/84 was the first decision to discuss disclosure in relation to a sub-range. The prior art in that case described a chemical process using a catalyst within a range from > 0 to <100 mol%. The claimed sub-range of the patent in suit was from 0,02 to 0,2 mol%.

The approach that only subject-matter described explicitly in the prior art documents had to be regarded as prejudicial to novelty, which was essentially normal practice in all patent offices, was not followed by the board, because the approach disregarded the purpose of Art. 54(1) EPC.[5]

The board pointed out ‘once again’ that the purpose of Art. 54(1) EPC is to prevent the state of the art from being patented again. Hence, the examination of novelty has to establish “whether the state of the art is likely to reveal the content of the invention’s subject-matter to the skilled person in a technical teaching”.[6] Applying this criterion, the board held that the extensive numerical range of the prior art “does not necessarily represent a disclosure[7], ruling out a selection from it, of all the numerical values between these minimum and maximum values if, as in this case, the sub-range selected is narrow and sufficiently far removed from the known range illustrated by means of examples.”

Hereby, the board introduced the new concepts of ‘narrow’ and ‘far removed’ for numerical sub-ranges, without specifying how to evaluate them. In the case in suit, ‘narrow’ and ‘far removed’ were determined on a purely numerical basis by comparing values of the prior art range and of the claimed sub-range[8].

The board showed that it applied the same novelty test for numerical sub-ranges as was applied earlier for chemical substances defined in a formula. If the prior art definition covers nine chemical substances of which only one is individualized by a disclosed embodiment, the other eight substances are not individualised and do not belong the prior art[9].

The board added a third criterion for novelty of a sub-range to prevent that the sub-range is not more than just a formal delimitation vis-à-vis the prior art, i.e. just an arbitrary selection from the prior art range. The board held that a selection is arbitrary if the sub-range has the same properties and capabilities as the prior art range. Novelty requires that the sub-range is a purposive selection in that an effect occurs within the sub-range that does not occur over the whole prior art range. The board emphasises that the sub-range is not new by virtue of the newly discovered effect occurring in it but must be new per se.[10]

2.2 Formulation of novelty test for sub-ranges

Although the novelty test for subranges was formulated by the above board in 1985, the test was included in the Guidelines only in 2005. The wording of the test in the Guidelines remained the same from 2005 till 2018:

“A sub-range selected from a broader numerical range of the prior art is considered novel, if each of the following three criteria is satisfied (see T 198/84 and T 279/89):

The end-points in the second criterion were not mentioned in the novelty tests of T124/87 and T279/89[12]. See the discussion in section 4 below.

The following sections discuss recent developments in case law that may affect the validity of the three criteria (a) – (c).

3 Purposive selection

The board of T198/84 “adheres to the principle that the sub-range singled out of a larger range is new not by virtue of a newly discovered effect occurring within it, but must be new per se” and “An effect of this kind is not therefore a prerequisite for novelty”.[13] Nevertheless, the board did include the new effect in the third criterion of the novelty test. Since the new effect is introduced to avoid an arbitrary selection and not to establish novelty, the third criterion is apparently inappropriate for a novelty test; its presence in a novelty test therefore lacks legal basis.

Recent decision 261/15 held, that the third criterion on purposive selection considers the presence of an effect of the claimed invention, which goes beyond the comparison of the claimed invention with the disclosure of the prior art. Therefore, whether the selection of the sub-range is purposive or not is more a question of inventive step than of novelty.[14]

The draft 2019 edition of the Guidelines that has entered into force on 01.11.2019 refers to T261/15 and dropped the third criterion on purposive selection.[15] Departments of first instance should therefore no longer use the third criterion when assessing the novelty of a sub-range[16].

However, the latest edition of the Case Law book, published in July 2019, mentions the three criteria and cites a few recent decisions supporting the third criterion in the novelty test, but in addition it also mentions a large body of case law issued over the last ten years that considers the third criterion an inventive step issue[17]. The boards have hereby caused a prolonged legal uncertainty in that it depends on the board handling the case whether or not the third criterion will be used[18].

4 End-points

It is established case law, that the disclosure of a range is an explicit disclosure of the end-points of the range.[19] This case law could have been the reason for including the end-points in the novelty test for sub-ranges in the Guidelines, as mentioned in the above section 2.2. Hence, the ‘far removed’ criterion requires that a sub-range must be far removed not only from disclosed examples but also from the end-points of a known range.[20]

Decision T261/15 held, that the criterion that a sub-range must be sufficiently far removed from the end-points of a known range, although mentioned in the Guidelines, is not stated in any jurisprudence, in particular not in the two decisions T198/84 and T279/89 cited in the passage of the Guidelines. In the view of the board, “the limit values of a known range, although explicitly disclosed, are not to be treated in the same way as the examples. The person skilled in the art would not, in the absence of further teaching in this direction, necessarily contemplate working in the region of the end-points of the prior art range, which are normally not representative of the gist of the prior art teaching”.[21] [22]

According to T261/15, a sub-range close to an end-point would not lack novelty because of the ‘far removed’ criterion. T279/89 appears to follow the same reasoning by not acknowledging novelty of a sub-range that is not near the lower or upper end of a known range but right in the middle thereof.[23]

It could be argued contrary to T261/15, that the teaching of a prior art patent document is sufficient for the skilled person to carry out the teaching in the sense of Art. 83 EPC also near the end-points and a technical effect of the disclosed invention is achievable throughout the entire range.[24] Hence, normally a skilled person would contemplate working anywhere in the range, including near the end-points, unless further teaching points away from the end-points.[25]

Another argument against T261/15 is, that the decision creates different types of disclosure: disclosures that can be used and disclosures that cannot be used for the novelty test for sub-ranges[26]. If prior art discloses a feature in the sense of making it available according to Art. 54(2) EPC and complying with the disclosure test, that feature should be usable in any novelty test under Art. 54(1) EPC[27].
All editions of the Guidelines since 2005, including the November 2019 edition, mention the end-points in the second criterion. In contrast, the Case Law book has never included the end-points in the second criterion[28]. The different treatment of end-points by departments of first and second instance causes legal uncertainty. For example, a sub-range near an end-point may be regarded not novel in first instance and novel in second instance.

5 ‘Narrow’ and ‘far removed’ and technology

The ‘narrow’ and ‘far removed’ criteria require that a sub-range is narrow compared to the known range and is far removed from known examples within the range. Since the assessment of novelty involves a comparison of a technical invention with the disclosure of technical prior art, the comparison must be based on technical arguments.

Nevertheless, the large majority of decisions on novelty of a sub-range assess compliance with the ‘narrow’ and ‘far removed’ criteria merely by an arbitrary comparison of just numerical values, devoid of any relation to technology. An example of such a decision is T1233/05, in which the known range is from 0 to 100% including an example having at least 50% and the claimed sub-range from 1 to 20%[29]. The sub-range ‘is narrow’ and, since 50% is ‘well above’ 20%, the range is ‘far away’ from the disclosed value. Hence, the board regarded the sub-range as new without addressing any technical issues.

T673/12 is one of the few decisions that use among others technical arguments in the assessment of ‘far removed’[30]. A product having a copper content in the range 0,6 – 2,7% and an example having 0,92% were known from a prior art document; a sub-range of 0,6 – 0,71% with favourable mechanical properties was claimed. The board regarded the example of 0,92% as not ‘far removed’ from the sub-range because 1) the example and the sub-range are both in the lower quarter of the known range (non-technical); 2) the example has a favourable mechanical strength (technical); 3) there is no teaching in the prior art document excluding a copper content in the sub-range from the teaching (technical). According to the board, the technical teaching for the assessment must be taken from the prior art document, not from the patent. Whether the skilled person is encouraged to use a copper content in the sub-range does not play a role in the assessment; it is rather whether the disclosure constitutes a usable technical teaching, i.e. whether the teaching is sufficiently supported by the examples.[31] [32]

The assessment of the ‘narrow’ and ‘far removed’ criteria based on numerical values without a relation to technology disqualifies the two criteria as novelty criteria.

A further problem with the ‘narrow’ criterion is, that the criterion itself is not related to technology. If the ‘narrow’ criterion is intended to prevent a sub-range from claiming a major part of a known range, thereby essentially just copying the known invention, it had better be part of the assessment of inventive step than of novelty.

Another difficulty of the ‘narrow’ and ‘far removed’ criteria is that the criteria are not clear in what ‘narrow’ and ‘far removed’ is. Is a sub-range of 40% of the known range narrow, or should it be 20%? The vagueness of the criteria set up in 1985 appears to be incompatible with the strictness of the requirement for novelty developed after 1990: for an invention to lack novelty, its subject-matter must be directly and unambiguously derivable from the prior art[33].

6 ‘Narrow’ and ‘far removed’ and disclosure of prior art

According to the board that introduced the three novelty criteria for sub-ranges, the ‘narrow’ and ‘far removed’ criteria establish, “whether the state of the art is likely to reveal the content of the invention’s subject-matter to the skilled person in a technical teaching” [34], in other words, whether “the person skilled in the art would seriously contemplate working” in the claimed sub-range[35].

The formulation of these novelty criteria date back to 1985. Since then, case law has further developed the concepts of novelty and disclosure. Decision T1085/13[36] summarises case law developed by the Enlarged Board of Appeal relating to novelty and disclosure of ranges, in particular decisions G2/88 and G2/10. It mentions the following relevant requirements for novelty.

It is emphasised that according to the Enlarged Board of Appeal a sub-range is not novel only if there is an implicit and / or explicit prior art disclosure within the sub-range. There are no other ways of taking away the novelty of the sub-range. The discussion below will focus on an implicit disclosure within the sub-range and not consider the obvious case of an explicit disclosure within the sub-range.
The present section investigates whether the two novelty criteria of 1985 still comply with above, recent interpretation of novelty and disclosure.

6.1 Clear and unambiguous

The vagueness in the phrases ‘likely to reveal’ in T198/84 and ‘would seriously contemplate’ in T261/15 are hardly compatible with the stricter requirement that the disclosure must be clear and unambiguous, i.e. admitting one interpretation.

Both phrases appear to interpret the word ‘available’ in Art. 54(2) EPC as using considerations of the skilled person to supplement the disclosure instead of using the considerations of the skilled person only to understand the disclosure[37]. In contrast, the Travaux Préparatoires held that the word ‘available’ emphasises the possibility that the public can take note of the invention[38].

If not only the interpretation by a skilled person but also the work of a skilled person is taken into consideration when assessing a prior art disclosure, the result is likely to be a further development of the prior art. Such a development will unavoidably include features that have not been disclosed in the prior art. That contravenes the requirement, that all the technical features of the claimed invention in combination must have been communicated to the public. Such a further development should be carried out under inventive step and not under novelty.[39]

6.2 ‘Narrow’ and disclosure

The first novelty criterion requires that a sub-range is narrow in view of the known range. If a sub-range is broader than the narrow range, it becomes not novel according to the criterion. Following the above requirement 6a), there must be an implicit prior art disclosure in this broader sub-range to take away its novelty. The implicit disclosure must be in the area where the broader sub-range extends beyond the narrow sub-range.

Since the criterion of ‘narrow’ is not linked to a specific location within the known range, the area where the implicit disclosure should be located to take away the novelty of the broader sub-range can be anywhere within the known range. Such an undetermined requirement is incompatible with the ‘unambiguous’ requirement 6b).

The only possible assessment of the criterion left is to compare just numerical values without regard to technology, as is usually done by the boards[40]. This disqualifies the first criterion as novelty criterion.

6.3 ‘Far removed’ and disclosure

The second novelty criterion requires that a sub-range is far removed from examples in the known, broad range. If a sub-range is not far removed but close to an example, it becomes not novel according to the criterion.

Applying the above novelty requirement 6a) to this situation, a sub-range close to an example is not novel because there is an implicit prior art disclosure within that sub-range. In other words, the sub-range is not novel if the teaching of an example having a parameter value outside the sub-range can be extended to an implicit disclosure having a parameter value within the sub-range.
Does a prior art document disclosing explicitly a range with one or more examples make such an implicit disclosure available? There are three approaches to what prior art discloses, as set out in the following three sub-sections.

6.3.1 Explicit – plus approach

The second criterion implies, that a known range with one or more explicitly mentioned examples in a prior art document not only discloses the explicitly mentioned examples but also a region around each example, in which a skilled person ‘would seriously contemplate working’.[41] Taking into account not only the explicitly disclosed examples but also a region around each example is called in this article the ‘explicit – plus approach’.

If the sub-range does not overlap a region, the sub-range is ‘far removed’ in terms of the second criterion and therefore novel; if the sub-range does overlap a region, it lacks novelty. To accord the second criterion with the novelty requirement 6a), there must be an implicit disclosure within the overlap of the sub-range and the region.

The implicit disclosure in the ‘far removed’ criterion appears not to comply with the requirements for novelty mentioned in the above section 6b) and 6c). The implicit disclosure falling within the sub-range includes a parameter value within the sub-range that is not explicitly mentioned in the prior art of the cases discussed. The prior art neither mentions a process that inevitably results in a parameter value within the sub-range. Hence, the parameter value within the sub-range is not an implicit disclosure of the prior art. Instead, the parameter value is a development of the prior art starting from the parameter value of an explicitly disclosed example. Such a development is more a matter of inventive step than of novelty.

There is another reason why an implicit disclosure as required for the ‘far removed’ criterion is not in agreement with case law. The EBoA held that the same disclosure test must be used for novelty and amendments.[42] If an implicit disclosure as required for the second criterion of the novelty test may be based on a known range with one or more explicit examples in a published patent application, the same implicit disclosure would be permitted as basis for an amendment under Art. 123(2) EPC of the patent application or patent. However, a disclosure of a range in general does not specifically and thus directly and unambiguously disclose all values within the range for the purpose of amending a range.[43] This must also apply to the assessment of novelty according to the EBoA and, consequently, to the ‘far removed’ criterion. Hence, the implicit disclosure used for the ‘far removed’ criterion is not in agreement with the disclosure test and is therefore not a true implicit disclosure.
Hence, the use of the ‘explicit – plus’ approach for the ‘far removed’ criterion is not in line with recent case law on availability.

6.3.2 Whole content approach
Decision T279/89 is usually quoted for its summary of the novelty test for sub-ranges[44] but not for its arguments on disclosure[45]. Since the novelty test is based on disclosure, it is worthwhile to consider its arguments.

The board cites approvingly T26/85:

“… what is made available to the public by means of a written document should not be restricted to the explicit disclosure, but extends to the whole content, i.e. to the information actually given to the person skilled in the art. When that information is sufficient to enable the skilled man to practice the technical teaching which is the subject-matter of the disclosure, taking into account also the general knowledge in the field to be expected of him, novelty can no longer be acknowledged.”

and finds further support for this whole content approach in T12/81[46] and T124/87[47]. The latter decision held that a prior art disclosure of a process for preparation of polymers is clearly not limited to the particular polymers, whose preparation is explicitly exemplified, but extends to the general class of polymers in the disclosure. All members of this general class are therefore available in the sense of Art. 54(2) EPC.

Decisions T12/81, T26/85 and T124/87 relate the disclosure of a prior art document and, hence, the availability of Art. 54(2) EPC to the sufficiency of disclosure of Art. 83 EPC.

As an exception to the disclosure of the whole content of a broad range, a sub-range of a broad range is not available if the skilled person is dissuaded from carrying out the invention in the sub-range, thereby making the sub-range novel[48].

It is a reasonable assumption that a range disclosed in a prior art patent document is normally sufficiently disclosed for the skilled person to work the invention over the entire range, because such a disclosure is required for a patent document by Art. 83 EPC. Dissuading statements not to use a part of a range are exceptional. The whole content approach for determining the disclosure of prior art would make it next to impossible to patent sub-ranges, because the novelty of the sub-range must be based on such an exceptional dissuading statement.

Case law on amendments does not support the whole content approach. As already mentioned in the above sub-section 6.3.1, a disclosure of a range in general does not specifically, and thus directly and unambiguously disclose all values within the range for the purpose of amending a range.[49]

Most decisions relating to sub-ranges do not use the whole content approach and assess compliance of a sub-range with the ‘far removed’ criterion by a mere comparison of numerical parameter values. Where a decision advances technical arguments in the assessment of the ‘far removed’ criterion, the reasoning often moves from the explicit-plus towards the whole content approach, without explaining why this approach can be used in the novelty test and not in the assessment of the allowability of amendments, whereas both should be based on the same disclosure test.[50]

6.3.3 Explicit-implicit approach

A further possibility is the explicit approach: a sub-range lacks novelty only if there is a prior art disclosure falling within the sub-range. The disclosure can be explicit or implicit, the latter in the sense of T1085/13 as set out in the above section 6. An advantage of this approach is that the assessment is straightforward and complies with the decisions of the Enlarged Board of Appeal about disclosure. Moreover, this interpretation of disclosure is essentially the normal practice in all patent offices.[51]

The author expresses his gratitude to Roel van Woudenberg and Cees Mulder for stimulating discussions and reviewing this article.

1. Travaux Préparatoires EPC 1973, Art. 54 E, document IV/2767/61-E, page 12, second paragraph. See also G2/88, reason 10.
2. G2/10, reason 4.6. The disclosure test is sometimes referred to as the ‘gold standard’; see G1/16, reason 17.
3. Reference is made to Roel van Woudenberg’s article in this issue of epi Information. The article is on the same topic but expresses a different point of view on several aspects.
4. T990/96, headnote
5. T198/84, reason 4, first sentence, second part. Decision issued on 28.02.1985.
6. T198/84, reason 4, last two sentences.
7. The board did not explain why the range disclosed in a prior art patent document is not a disclosure of all numerical values between the end-points whereas the same prior art is a sufficient disclosure in the sense of Art. 83 EPC allowing the skilled person to carry out the invention for each numerical value within the range. If the disclosure had been insufficient, the prior art document should not have been used as prior art (see e.g. T1437/07, reason 25 and T1457/09 reason 36). See section 6.3.2 below.
8. T198/84, reason 5. The sub-range 0.02 to 0.2 is a small specimen from the prior art range >0 to <100 and removed at least a power of ten from the prior art embodiments between 2 to 13. Note, that the board does not refer to the very small distance between the sub-range and the lower end-point of the prior art range, only to the distance between the sub-range and disclosed examples.
9. T198/84, reason 6, supporting the reason by a reference to T181/82.
10. T198/84, reason 7, supporting the reason by a reference to T12/81. A peculiarity of the third criterion is that, although the presence of a new effect does not make a sub-range novel, lack of a new effect does make the sub-range not novel.
11. Guidelines for Examination in the EPO, edition November 2018, G-VI, 8(ii)
12. T261/15 reason 2.3.2
13. T198/84, reason 7. Likewise, T230/07, headnote, states “A sub-range is not rendered novel by virtue of a newly discovered effect occurring within it.”
14. T261/15, reason 2.2.2. Confirmed by a large body of case law; see the book Case Law of the Boards of Appeal of the European Patent Office, 9^th edition 2019 (CLBA) I.C.6.3.1.
15. Draft Guidelines, edition 2019, G-VI, 8(ii), published on the EPO website on 01.10.2019. The published draft still mentions in the paragraph below the three criteria the relevance of a technical effect in the sub-range for novelty, which appears to be inconsistent with the deletion of the purposive selection requirement.
16. As a consequence, a patent granted on three criteria may, after entry into force of the Guidelines 2019 on 01.11.2019, be assessed in opposition on two criteria and subsequently in appeal on three criteria.
17. CLBA I.C.6.3.1
18. As a consequence, a patent application refused on three criteria before 01.11.2019 may be re-assessed by a board using two criteria, or a patent application refused on two criteria after 01.11.2019 may be re-assessed by a board using two criteria.
19. T1115/09, reason 2.3, cited in CLBA IC.6.3.2.
20. The mention of the end-points of the known range in the ‘far removed‘ criterion of the Guidelines would probably reverse decision T198/84 by making the sub-range discussed in the above section 2.1 not novel (see footnote 7).
21. T261/15, reason 2.3.2, one but last paragraph.
22. The gist of the prior art teaching, used for the special treatment of the end-points, normally relates to the attainment of a technical effect. Hence, a technical effect is used to disregard end-points in the second criterion. The same decision, T261/15, refuses to use a technical effect in the third criterion (see the above section 3). The decision does not reconcile these apparently different treatments of technical effect in the novelty assessment.
23. T279/89, reason 4.1.
24. CLBA I.D.9.8.3
25. Followed by e.g. T673/12, reason 2.3
26. Although, according to the decision, an end-point should not be used for the second criterion, an end-point can be used as a disclosure to destroy the novelty a claimed range (Guidelines G-VI, 8(iii)) and be used as basis for amending a range (CLBA II.E.1.5.1).
27. G1/16, reason 17, mentioning the applicability of the disclosure test to novelty.
28. This applies to all editions of Case Law of the BoA, including the 9^th edition of July 2019.
29. T1233/05, reason 4.3. Another example is T230/07, reason 4.1.5.
30. T673/12, reason 2.3
31. Note, that the reasoning of the board about disclosure is similar to that in T279/89, reason 4.2, which made the patentability of sub-ranges next to impossible. See section 6.3.2 below.
32. The author has not found any decision using technical arguments in the assessment of ‘narrow’.
33. See the disclosure test in the above section 1.
34. T198/84, reason 4, last two sentences.
35. T261/15, reason 2.2.2. In contrast, ‘seriously contemplating’ is not necessary according to T673/12, reason 2.3.
36. T1085/13, reason 3.6. The decision, issued in 2018, held that the well-established novelty test for purity of a compound was not in line with G2/88 and G2/10
37. See T1085/13, reason 3.6.4.
38. See the above section 1 and the requirement c) in the above section 6.
39. See e.g. T1085/13, reason 3.6 – 3.10
40. See the above section 5.
41. T261/15, reason 2.2.2
42. G2/10, reason 4.6; CLBA II.E.1.1
43. T985/06 reason 2.1.2; CLBA II.E.1.5.3
44. T279/89, reason 4.1. See the above section 2.2.
45. T279/89, reason 4.2 – 4.5. Reason 4.5 states that use of whole contents approach, which uses a broader definition of the concept of novelty than the explicit disclosure approach, is constant jurisprudence of the boards and established practice of the EPO. In spite of this assertion of the board, its arguments on availability are not cited in the Guidelines nor in the book Case Law of the Boards of Appeal.
46. T12/81, reasons 5, 7 – 9
47. T124/87, reason 3.4
48. T279/89, reason 4.2 for a sub-range entirely within a known range; T26/85, headnote, for an overlapping range.
49. T985/06 reason 2.1.2 and 2.1.4; CLBA II.E.1.5.3
50. See e.g. T261/15, reason 2.2.2 and 2.3.2; T673/12, reason 2.3
51. T198/84, reason 4, second part of first sentence

Comments